,
Mechanism of degradation of the steroid side
chain in the formation of bile acids
Ingemar Bjorkhem
Department of Clinical Chemistry, Karolinska Institute, Huddinge Hospital, Huddinge, Sweden
Introduction
The 27-Hydroxylase Pathway
27-Hydroxylation
Conversion of 27-hydroxysteroid to 27carboxysteroid
Peroxisomal Boxidation of coprostanoic acids
The 25Hydroxylase Pathway
Relative Importance of the 25- and 27-Hydroxylase Pathways
Species Differences and Alternative Substrates for the
27-Hydroxylase Pathway
Pathway involving 7a,27-dihydroxy-4-cholesten-3-oneas
intermediate
Pathway involving 7a,12a,27-trihydroxy4cholesten-3-one
as intermediate
Pathway involving 27-hydroxycholesteroI as intermediate
Inborn Errors in Side-Chain Cleavage
Cerebrotendinous xanthomatosis
Peroxisomal disorders
Mechanism of Degradation of the Steroid Side Chain of
Plant Steroids
Concluding Remarks
INTRODUCTION
The most important pathway for the metabolism
and excretion of cholesterol in mammals is the forma-
tion of bile acids. The two major primary bile acids,
cholic and chenodeoxycholic acids, are formed in the
liver and secreted in bile to the intestine.
The conversion of cholesterol into bile acids invol-
ves almost all the conceivable mechanisms for conver-
sion of a lipophilic compound into an excretable
water-soluble product (for a general review, see ref. 1).
Of the more than 15 different enzymes participating
in the conversion, hydroxylases, oxidoreductases, and
conjugating systems are of particular importance for
increasing the polarity. The enzymes that modify the
steroid nucleus are able to convert the nonpolar
3~hydroxyA5-steroidinto a considerably more polar
5~holestane-?i~~,7~~-dihydroxy- or 5pcholestane-Sa,
7a,12a-trihydroxy steroid. The enzymes involved in
these conversions are mainly located in the endoplas-
mic reticulum and the cytosol. The enzymes involved
in the steroid side-chain degradation convert the high-
ly nonpolar Cpsteroid side-chain into a chain-shor-
tened carboxylic acid conjugated to an amino acid.
These enzymes are mainly located in the mitochondria
and in the peroxisomes.
According to current concepts, the conversion of
cholesterol into bile acids in mammals starts with the
Downloaded from www.jlr.org by guest, on May 14, 2019
nuclear transformations, and most or all of these chan-
ges precede those of the steroid side-chain. Based on
early in vivo and in vitro work on rats, the sequence of
reactions shown in Fig. 1 was formulated about 25
years ago for the conversion of cholesterol into cholic
and chenodeoxycholic acids. It is evident, however,
that alternative pathways exist where some or most of
the changes in the steroid side-chain precede the
changes in the nucleus. As the flux of bile acids
probably does not regulate any of the pathways where
7a-hydroxylation is not the first step, it seems that the
cholesterol 7a-hydroxylase is the only enzyme that is
capable of regulating the overall conversion into bile
acids.
The steroid side chain of plant sterols contains a
methyl or an ethyl group in 24position. This means
that the normal mechanism for side-chain degradation
cannot be operative. In view of the very low degree of
absorption of these steroids, there is less need for a
specialized degradative enzyme. In at least one mam-
malian species, however, a system has been evolved
that is able to convert plant sterols into highly polar
conjugated Cnl-bile acids.
In the present review, the emphasis is on the most
important mechanism for steroid side-chain degrada-
tion in connection with bile acid biosynthesis, the 27-
Abbreviations: THCA, trihydroxycoprostanoic acid; DHCA,
dihydroxy-5~holestanoicacid (dihydroxycoprostanoic acid) ; CTX,
cerebrotendinous xanthomatosis.
Journal of Lipid Research Volume 33, 1992 455
 HO
i
I
t '0H

0ai? 0
1

Downloaded from www.jlr.org by guest, on May 14, 2019

HO' HO'

CH,OH
t
CH20H

HO
HO'
I 1

H0'
+ HO'Z C O O H

HO'&OoH
HO' zoo.
Fig. 1. Major pathway for the normal biosynthesis of bile acids in the rat.

hydroxylase pathway. An alternative pathway that THE 27-HYDROXYLASE PATHWAY

operates when the major pathway is blocked will also
be reviewed. Alternative substrates for sidechain
degradation in different species, as well as inborn er-
rors of metabolism affecting side-chain degradation,
are also discussed. A recently discovered new pathway
for degradation of the steroid sidechain of plant
sterols will be mentioned.
It is well established that the major mechanism for
degradation of the cholesterol sidechain starts with
oxidation of one of the two terminal methyl groups
(ohydroxylation) followed by fhxidation and con-
jugation. A simplified scheme of the reactions is shown
in Fig. 2.

456 Journal of Lipid Research Volume 33, 1992

 Downloaded from www.jlr.org by guest, on May 14, 2019
m R CO.CoA + <Co.CoA

. NHCH,COOH

HO’

Fig. 2. Side-chain oxidation and conjugation in connection with bile acid biosynthesis from cholesterol.
The 27-hydroxylase pathway for the conversion of 5&cholestane-3a,7a,12a-triol
into cholic acid is shown.

The above pathway will be referred to as the “W
hydroxylase pathway” or the “27-hydroxylasepathway.”
27-Hydroxylation
Inasmuch as hydroxylation of one of the two ter-
minal methyl groups in the steroid sidechain creates
an asymetric carbon atom at C25, @hydroxylation
may be stereospecific. Popjak et al. (2) have suggested
that the 25-pro-R methyl group should be denoted C-
26 and the 25-pro-S methyl group C-27. There is a
mitochondrial @hydroxylase in mammalian liver that
almost exclusively attacks the C-27 methyl group (for a
general review, see ref. 1 ) . This hydroxylase should
thus be denoted “27-hydroxylase,” although it has
been referred to as “26hydroxylase” in most previous
work. There is also a microsomal *hydroxylase in
mammalian liver, which has a specificity towards the C-
26 methyl group (3). In humans, this microsomal 2 6
hydroxylase has little activity compared to the
mitochondrial 27-hydroxylase (4).In accordance with
this, trihydroxycoprostanoic acid isolated from human
bile has the 25-R-configuration (5), indicating that it is

Bj6rkhem Mechanism of degradation of steroid side chain in formation of bile acids 457
formed by an initial attack by the mitochondrial 27-
hydroxylase. In addition to 26-hydroxylase activity,
both rat and human liver microsomes contain 2 3 , 2 4 ,
and 25-hydroxylases active towards the Cnrsteroid side-
chain (cf below). In the rat, the microsomal 26-
hydroxylase may be of importance and there is some
evidence that in this species the enzyme may be in-
volved in the regulation of the ratio between cholic
acid and chenodeoxycholic acid formed in the liver
(6).
The microsomal 26-hydroxylase has a high substrate
specificity and the mitochondrial 27-hydroxylase has a
low substrate specificity (see below). The mitochon-
drial 27-hydroxylase is located not only in the liver but
also in fibroblasts (7), brain (S), kidney (9), and
several other organs and tissues (10).
On the basis of experiments with rat liver
peroxisomes, it has been suggested that there is an w
hydroxylase in peroxisomes active toward Cz;-steroids
(11).Attempts to confirm this in our laboratory and in
the laboratory of Pedersen have failed, however.
In view of the lesser importance of the 2t%
hydroxylase in mammals, only the 27-hydroxylase will
be discussed in this review. Early work by Bjdrkhem
and Gustafsson (12, 13) and by Taniguchi, Hoshita,
and Okuda (14) and Okuda, Weber, and Ullrich (15)
established that this enzyme is a mixed function
oxidase containing cytochrome P-450. The enzyme is
bound to the inner mitochondrial membrane. Thus
the activity was low with intact mitochondria and
NADPH as a co-factor (13, 14). Under such conditions
citric acid and isocitric acid, which are able to
penetrate the inner mitochondrial membranes, stimu-
late 27-hydroxylation much more efficiently than
NADPH (13, 14). It is evident that citric acid and
isocitric acid generate NADPH inside the mitochon-
drial membranes. With leaking mitochondria, NADPH
stimulates the reaction as efficiently as isocitrate.
There is some evidence for the presence of different
transport mechanisms for different substrates through
the mitochondrial membranes (16). This may be of
some importance for the different rates of hydroxyla-
tion of different substrates by intact liver
mitochondria.
Pedersen, Oftebro, and Vinngird (17) and Sato et
al. (18) reported simultaneously that small amounts of
cytochrome P-450 could be solubilized from the inner
membranes of rat liver mitochondria. The cytochrome
P-450 fraction obtained was active toward cholesterol
as well as toward 5@holestane-W,7a, 12a-triol in the
presence of ferredoxin, ferredoxin reductase, and
NADPH. Ferredoxin and ferredoxin reductase were
active regardless of whether they were isolated from
rat liver mitochondria or bovine adrenal mitochondria
(19). Pedersen, Bjdrkhem, and Gustafsson (19)
458 Journal of Lipid Research Volume 33, 1992
showed that the substrate specificity of a reconstituted
system from rat liver mitochondria was similar to that
in intact mitochondria. The rate of 27-hydroxylation
was, however, about 10-fold higher in the reconstituted
system than in intact mitochondria, indicating that the
transport of the steroid through the mitochondria
membranes may be a limiting factor (19).
Wikvall (20) and Dahlback and Wikvall (21) were
the first to purify the mitochondrial 27-hydroxylase to
homogeneity. They used rabbit liver as enzyme source.
Later, other groups characterized the rat (22), human
(23), and pig (24) 27-hydroxylases. Andersson et al.
(10) recently made an extensive characterization of
the rabbit liver 27-hydroxylase. The protein sequence
of the 499 amino acid enzyme shares many of the
hallmarks of a cytochrome P-450 structure, including
an overall hydrophobic nature (36% of the amino
acids in this protein have either aromatic or
hydrophobic sideshains) and a conserved cystein
residue (located at position 444 in the 27-hydroxylase)
that is thought to be a ligand for the heme iron. The
Downloaded from www.jlr.org by guest, on May 14, 2019
structure of the enzyme displays a similarity with the
mitochondrial cholesterol sidechain cleavage enzyme
(34%) and to the adrenal mitochondrial 1lphydroxyl-
ase (33%).A signal sequence of 36 residues was found
to precede the coding region and this signal sequence
is common to sequences that direct proteins into the
mitochondrion. The 27-hydroxylase was regarded as a
member of a novel cytochrome P-450 gene family.
Several research groups in the bile acid field have now
decided to call the gene CW 27.
Andersson et al. (10) were also able to express the
27-hydroxylase activity in cultured monkey COS cells.
27-Hydroxylase enzyme activity in COSM6 cells trans-
fected with the cDNA was enhanced by co-transfection
of a plasmid encoding bovine adrenodoxin but not by
a plasmid encoding NADPH cytochrome P-450 reduc-
tase. As could have been expected from previous work,
mRNA measurements showed that the enzyme was ex-
pressed in several different organs and tissues (10).
Most interesting, the abundance of the mRNA for the
enzyme paralleled the cholesterol biosynthesis capacity
of the tissues that were assayed. Thus the mRNA levels
were high in the liver, duodenum, adrenal gland, and
lung and less abundant in the kidney and spleen.
The liver mitochondrial 27-hydroxylase seems to be
of little importance for the regulation of bile acid
biosynthesis and the composition of bile acids formed.
The activity is high compared to that of the enzyme
that catalyzes the rate-limiting step, cholesterol 7a-
hydroxylase. It has been reported that the rat enzyme
is slightly inhibited by biliary drainage and that it is
not affected by, for example, starvation and phenobar-
bital treatment (12). Biliary obstruction seems to have
different effects on the rat mitochondrial 27-hydroxyl-
ase, depending on the substrate (25). Whether these
effects are due to effects on the 27-hydroxylase per se
or to effects on the transfer of steroids to the site of
the enzyme could not be established. Saarem and
Pedersen (26) was well as Dahlbick (27) reported that
the level of the enzyme is higher in female than in
male animals. Cholic acid and starvation was found to
reduce the amount of mitochondrial cytochrome P-
450-27 and the catalytic activity by 30-60%. There were
no parallel changes in the mRNA encoding the en-
zyme however (27).
The rat, human, and pig 27-hydroxylases have
catalytical properties similar to those of the rabbit en-
zyme (22-24). The deduced amino acid sequences of
the rat and the human enzymes indicated that it is
73% and 8l%, respectively, identical to the rabbit en-
zyme (23, 28). It may be mentioned in this connection
that Raza and Avadhani (29) isolated two bnaph-
thoflavone-inducible species of cytochrome P-450 from
rat liver that possessed 27-hydroxylase activity in addi-
tion to hydroxylase activity toward various xenobiotics.
In contrast to the cytochrome P-450 fractions
described above, the preparation studied by Raza and
Avadhani (29) could be reconstituted with both
microsomal NADPH-cytochrome P-450 reductase and
mitochondrial ferredoxin + ferredoxin reductase.
The possibility has been discussed that the 27-
hydroxylase may be of importance for the overall
regulation of cholesterol biosynthesis (for a review, see
ref. 30). One of the products, 27-hydroxycholesterol,is
thus known to be a potent inhibitor of the rate-limit-
ing enzyme in cholesterol synthesis, the HMGCoA
reductase (31). The finding by Andersson et al. (10)
that the abundance of the mRNA for the enzyme
paralleled the cholesterol biosynthesis capacity of the
tissues is also in agreement with this possibility. Since it
is possible to survive a complete or almost complete
lack of the enzyme activity for a life-time (cf below), it
is evident that the enzyme activity cannot be obligatory
for the down-regulation of cholesterol synthesis in
various tissues. In recent work from this laboratory, the
possibility was excluded that a mitochondrial 27-
hydroxylation is of importance in the down-regulation
of HMGCoA reductase by dietary cholesterol in mice
(E. Lund and I. Bjorkhem, unpublished results). It was
shown that 27-hydroxylation of 26,26,26,27,27,27-?H~-
cholesterol in mouse liver mitochondria was associated
with a marked isotope effect. When such deuterated
cholesterol was added to the diet, it was able to sup-
press HMGCoA reductase as efficiently as unlabeled
cholesterol.
It may be speculated that the 27-hydroxylase may be
involved in the transport of cholesterol out from the
cell. Thus 27-hydroxycholesterol and, in particular, its
oxidation product 3khydroxy-5cholestenoic acid are
Bjiirlthem Mechanism of degradation of steroid side chain in formation of bile acids
considerably more polar than cholesterol. Direct
evidence for this hypothesis is still lacking, however.
Pedersen, Holmberg, and Bjdrkhem (32) and
Bjorkhem et al. (33) showed that a crude preparation
of cytochrome P-450 from rat liver mitochondria was
able to catalyze not only 27-hydroxylation of various
C2rsteroids but also 25hydroxylation of cholesterol
and vitamin Ds. A kinetic study suggested that dif-
ferent enzymes, or at least different binding sites, may
be involved in the 25- and 27-hydroxylations (33). In
accordance with this contention, Dahlbick (34)
showed that monoclonar antibodies against a purified
27-hydroxylase inhibited 27-hydroxylation but not 25-
hydroxylation by a reconstituted system containing
highly purified, apparently homogeneous mitochon-
drial cytochrome P-450, ferredoxin, and ferredoxin
reductase. In contrast to these results, however,
Ohyama et al. (35) reported that partial denaturation
by heating and treatment of the enzyme by N-
bromosuccinimide inactivated the two enzyme ac-
tivities in a similar manner. It has also been shown that
Downloaded from www.jlr.org by guest, on May 14, 2019
COS-cells transfected with the 27-hydroxylase plasmid
expressed both 27- and 25-hydroxylase activity (27). At
least in these cells, the same gene must therefore be
responsible for both activities.
Post-translational changes in the primary gene
product or presence of a specific 27-hydroxylase in ad-
dition to the more nonspecific 27-hydroxylase (posses-
sing both 27- and 25-hydroxylase activity) may explain
the apparently conflicting results from the different
groups. Another possibility is that the binding of non-
polar substrates like cholesterol and vitamin D3 to the
active site of the enzyme may differ from the cor-
responding binding of more polar substrates like 5 k
cholestane-3a,7a,12a-triol.In accordance with this
latter hypothesis we have found that cyclosporin in-
hibits 27-hydroxylation (and 25-hydroxylation) of cho-
lesterol and 25-hydroxylation of vitamin D but not
27-hydroxylation of more polar Czrsteroids catalyzed
by an apparently homogeneous cytochrome P-450
preparation from rabbit liver (H. Dahlbick, I.
Bjorkhem, and H. Princen, unpublished observation).
Conversion of 27-hydroxysteroidto 27carboxysteroid
The liver has a very high capacity to catalyze oxida-
tion of 27-hydroxysteroids into the corresponding 27-
carboxysteroids. Three different systems have been
described.
In 1965 Okuda and Danielsson (36) synthesized the
27-aldehyde, believed to be the intermediate in the
above conversion. This aldehyde was efficiently oxidiz-
ed into 3a,7a,l2a-trihydroxy-5~-cholestanoic acid (tri-
hydroxycoprostanoic acid, (THCA) in the cytosolic,
microsomal, and mitochondrial fractions of a rat liver
homogenate. The soluble. and the mitochondrial frac-
459
tions were most efficient. The soluble 5bholestane-
3a,7~~,12~~,27-tetrol-NADdehydrogenase and 3a,7a,
12a-trihydroxy-5~cholestan-27-al-NAD-dehydrogenase be catalyzed by the mitochondrial fraction (45, 47, 48)
were later purified by Okuda and Takigawa (37-39). It
was suggested that the enzymes could be identical to al-
cohol and aldehyde dehydrogenase, respectively. In ac-
cordance with this, Waller, Theorell, and Sjovall (40)
and Bjorkhem, Jornwall, and Akesson (41) reported
that recrystallized alcohol dehydrogenase from horse
liver was able to catalyze oxidation of the tetrol into the
carboxylic acid. The SS isoenzyme was found to be most
efficient (41).
In accordance with the investigation by Okuda and
Danielsson (36), Dahlback et al. (42) reported that
rabbit liver mitochondria contain an NADdependent
enzyme system capable of catalyzing oxidation of the
tetrol into the carboxylic acid.
Anderson et al. (10) and Dahlback and Holmberg
(43) showed that the purified mitochondrial cyto-
chrome P-450-27 was able to catalyze formation of
3a,7a,l2a-trihydroxy-5~holestanoicacid from 5p- with different subcellular fractions of rat liver, the
cholestane-3a,7a,12al27-tetro1when reconstituted
with ferredoxin, ferredoxin reductase, and NADPH.
The possibility was clearly excluded that the enzymatic
activity was derived from contaminating mitochondrial
and cytosolic dehydrogenases. The intermediate al-
dehyde could not be isolated. It is possible, however,
that an intermediate aldehyde may be dismutated to
give equal amounts of the tetrol and the carboxylic
acid.
The relative importance of the three systems is not
known. Sjdvall, Anderson, and Lieber (44) showed
that deer mice, which genetically lack soluble alcohol
dehydrogenase, have the same bile acid pool size and
composition as animals with normal levels of the en-
zyme. It was therefore concluded that the soluble al-
cohol dehydrogenase cannot be obligatory for bile
acid biosynthesis.
Inasmuch as the 27-hydroxylation step occurs inside
the mitochondria, it is attractive to suggest that one or
both of the two mitochondrial systems are most impor-
tant for the conversion. Because the carboxylic acid is
considerably more polar than its precursor, it is pos-
sible that the conversion is of importance for the trans-
port of the bile acid intermediate from the
mitochondria.
Peroxisomal Boxidation of coprostanoic acids
Early experiments by Masui and Staple (45) showed
that 3a,7a,12a,24tetrahydroxy-5~holestanoic acid is deuterium
a probable intermediate in the conversion of THCA
into cholic acid and that the 24hydroxylation step was
catalyzed by the mitochondrial fraction of a rat liver
homogenate. It was subsequently shown that this
hydroxylation was also catalyzed by the microsomal
460 Journal of Lipid Research Volume 33, 1992
fraction together with the cytosol and ATP (46). The
final oxidation and thiolytic cleavage were reported to
but also by the soluble fraction (45, 47) and by the
microsomal fraction combined with the 100,000 g su-
pernatant (48). In these early experiments no attempts
were made to separate out other subcellular fractions
and marker enzymes were not used to check for cross-
contamination of the fractions.
In view of the controversial results, the subcellular
localization of the enzyme system responsible for the
overall conversion of THCA into cholic acid and of
DHCA (3a,7adihydroxy-5bholestanoic acid, dihy-
droxycoprostanoic acid) into chenodeoxycholic acid
was reinvestigated by Pedersen and Gustafsson (49)
and by Pedersen's group in collaboration with the
author (50, 51). The initial experiments showed that a
peroxisome-enriched fraction from rat liver had the
highest capacity to catalyze conversion of THCA into
cholic acid. When THCA and DHCA were incubated
Downloaded from www.jlr.org by guest, on May 14, 2019
highest specific activities for conversion into cholic
acid and chenodeoxycholic acid were observed in a
light mitochondrial fraction that also contained the
highest activity of the peroxisomal marker enzymes
(50, 51). After separation of the light mitochondrial
fraction on sucrose or Nycodenz gradients, it was
shown that the fractions containing the highest
peroxisomal marker activities also catalyzed formation
of cholic acid from THCA and chenodeoxycholic acid
from DHCA. The conversion required the presence of
NAD, CoA, ATP and Mg". With peroxisomes prepared
on sucrose gradient, FAD stimulated the reaction. The
reaction was stimulated by KCN and unaffected by in-
hibitors of the mitochondrial respiratory chain. Small
amounts of 24OH-THCA could be isolated from in-
cubations with THCA.
The findings were consistent with the view that the
reaction sequence for oxidative cleavage of the C y 7
steroid side-chain is similar to that of the peroxisomal
boxidation of fatty acids, involving a THCA-CoA syn-
thetase, an FADdependent oxidase, a hydratase, an
NADdependent dehydrogenase, and a thiolase (Fig.
2). A AZ4-unsaturated steroid should be an inter-
mediate in this reaction. Indirect evidence for the in-
termediate formation of such a compound was
obtained from experiments performed in the presence
of 'H20 or in an atmosphere of "Op (52). In the
former case there was an incorporation of one atom of
into the 25-position, whereas in the latter
case there was no incorporation of isotope. Using
specific incubation conditions without NAD, it was
later possible to show the direct conversion of THCA
into Az4-THCA (53). It was also shown that the AZ4-
double bond had the ZZ-configuration.
 The first step in the overall conversion of THCA
into cholic acid is the activation to yield a CoA deriva-
tive, The microsomal fraction of a rat liver homoge-
nate was found to have the highest catalytic activity,
whereas the peroxisomal fraction had little or no
capacity to catalyze this conversion (54). The THCA-
CoA oxidase was found to be a specific system, not
identical to that involved in peroxisomal fatty acid
oxidation. In contrast to the fatty acid oxidation sys-
tem, clofibrate did not stimulate the THCA oxidase
(55) and treatment of rats with partially hydrogenated
marine oil was found to have different effects on the
two systems (56). Schepers et al. (57) described a par-
tial purification of the THCA-CoA oxidase from rat
liver peroxisomes and showed that it was different
from palmitoyl-CoA oxidase. The THCA-CoA oxidase
had an apparent molecular mass of 139 kDa and con-
sisted mainly, if not exclusively, of one polypeptide
component of 69 kDa.
The next two steps in the conversion are catalyzed
by a bifunctional enzyme, enoyl-CoA hydratase/ khy-
droxyacyl-CoA dehydrogenase. This enzyme seems to
be the same as that involved in peroxisomal fatty acid
oxidation. Using photoaffinity labeling of isolated intact
rat liver peroxisomes with tritium-labeled 7,7-azo-
3a,12a-dihydroxy-5Pcholestan-27-oyl-CoA, Gengenbac-
her et al. (58) showed an incorporation of radioactivity
into polypeptides that immunoprecipitated with an-
tibodies toward both the bifunctional enzyme and the
thiolase involved in peroxisomal Poxidation of fatty
acids.
The final step in the sequence of reactions leading
to a conjugated bile acid is catalyzed by a bile acid-
CoA:amino acid N-acyltransferase. Kase et al. (59) and
Kase and Bjorkhem (60) showed that the highest
specific amidation activity of both choloyl-CoA and
chenodeoxycholoyl-CoA was always found in the most
peroxisome-rich fractions of a rat liver homogenate.
The microsomal fraction contained less activity, while
the cytosol was inactive. In previous work substantial
amounts of enzyme activity were found in the cytosol.
This may be due to leakage of peroxisomal enzymes
into the cytosol during the preparation of the different
subcellular fractions. Striking differences were o b
served in the & values and the saturation concentra-
tions for glycine and taurine in the peroxisomal
system. It was suggested that most of the de novo syn-
thesized bile acids conjugated to taurine by the
peroxisomal systems, whereas the bile acids decon-
jugated in the gut and recirculation to the liver might
be activated and amidated by the microsomal enzyme
system prior to biliary secretion.
Very recently Johnson et al. (61) isolated a bile acid-
CoA:amino acid N-acyltransferase from the soluble
fraction of homogenized purified frozen human liver.
Bjiirlzhem Mechanism of degradation of steroid side chain in formation of bile acids
After purification, the reduced denatured enzyme
migrated as a single 50 kDa protein band by sodium
dodecyl sulfate-polyacrylamide gel electrophoresis. A
similar molecular mass was obtained for the native en-
zyme by HPLC gel filtration. The purified enzyme was
found to use glycine, taurine, and 2-fluoro-&alanine
but not alanine as substrates. It was suggested that a
single monomeric enzyme is responsible for conjuga-
tion of bile acids with both glycine and taurine in
human liver.
THE 25-HYDROXYLASE PATHWAY
In addition to the 27-hydroxylase pathway, there is
an alternative mechanism for degradation of the
steroid sidechain in 5bholestane-3a,7a,lPa-triol. It
was shown early on in both rat (64) and human (4)
that 5bholestane-3a,7a, 12a-triol is efficiently 25-
hydroxylated in the microsomal fraction. Shefer et al.
(63) and Salen et al. (64) further demonstrated that
the product, 5~cholestane-3a,7a,12a,25-tetrol, is
Downloaded from www.jlr.org by guest, on May 14, 2019
converted into 5(%cholestane-3a,7a,l2a,24a,25-pentol,
5bholestane-3a,7a,12a24P,25-pentol, 5bholestane-
3a,7a,12a,23,25pentol, and 5~cholestane-3a,7a, 1213,
25,26-pentol in the presence of microsomes fortified
with NADPH. In the presence of NAD, 5bholestane-
3a,7a,12a,24P,25-pentol, but not the other 5Pcholes-
tanepentols formed, is converted into cholic acid by
soluble enzymes (Fig. 3). The latter conversion must
be assumed to involve acetone. These experiments
demonstrated the existence of a new pathway for side-
chain degradation in cholic acid biosynthesis that does
not involve hydroxylation at C-26 or the participation
of mitochondria. The relative importance of this path-
way has been a matter of controversy (cf below). Since
HO' c$-
,c$
l -OH-
I I1
I11 IV
Fig. 3. Formation of cholic acid from 5~holestane-3a,7a,l2a-triol
by the 25-hydroxylase pathway described by Salen et al. (64) I = 5p-
cholestane-3a,7a,l2a-triol; I1 = 5~holestane-Sa,7a,12a,25-tetrol;
111 = 5pcholestane-3a,7a,12a,24(S),25-pentol;IV = cholic acid.
461
there is little or no 25-hydroxylation of 5bholestane-
3a,7a-diol in liver microsomes, the 25-hydroxylase
pathway cannot be of importance in the formation of
chenodeoxycholic acid.
It may be mentioned that the microsomal 25-
HO O Y
hydroxylase is not active towards cholesterol (4). As
mentioned above, the mitochondrial fraction of a liver
homogenate is able to catalyze 25-hydroxylation of
27-hydroxylase

Pathwa/

HO014C CH,?4CH,
~
\ 25-hydroxylase
Pathway
C O / c14
H3

W H 3
cholesterol (13). The rate of this hydroxylation is, trapped a s
however, very low and it seems unlikely that a pathway 2.4-dinitrophenylhydrazone

involving 25-hydroxycholesterol is of importance in 14

CO2
the biosynthesis of bile acids. trapped with
phenethylamine

Fig. 4. Experimental approach used by Duane et al. (65, 66) to

RELATIVE IMPORTANCE OF THE 2 5 AND evaluate the relative importance of the 25- and the 27-hydroxylase
27-HYDROXYLASE PATHWAYS pathways in the biosynthesis of bile acids in rat and humans.

On the basis of the efficient 25-hydroxylation of 5k

cholestane-3a,7a,l2a-triolin human liver microsomes
and the accumulation of 5fkholestane-3a,7a,l2a,25-
side-chain. Thus 5~holestane-3a,7a-diolis the sub-
tetrol in bile and feces of patients with cerebroten-
strate for the 27-hydroxylase in the formation of
dinous xanthomatosis (CTX, see below), Salen et al.

Downloaded from www.jlr.org by guest, on May 14, 2019

chenodeoxycholic acid and 5bholestane-3a,7a,12a-
suggested that the 25-hydroxylase pathway may be the
triol is the corresponding substrate in the formation of
major mechanism for biosynthesis of cholic acid in
cholic acid. This pathway is most probably
humans (64). There are, however, a number of experi-
predominant in rats but the situation may be different
mental findings that suggest that the 27-hydroxylase
in humans.
pathway is most important in the biosynthesis of both
Since the mitochondrial 27-hydroxylase has a broad
cholic acid and chenodeoxycholic acid.
substrate specificity (4, 12, 13), it is evident that path-
First, there are two known inborn errors of metabo-
ways may exist where the 27-hydroxyl group is intro-
lism in the 27-hydroxylase pathway in humans (cf
duced at a stage where only part of the nuclear
below). In both cases the metabolic block leads to ac-
changes have occurred. In the extreme case cholester-
cumulation of intermediates (or their metabolites) in
ol may be 27-hydroxylated prior to the introduction of
the 27-hydroxylase pathway. In one of the diseases
the 7a-hydroxyl group. Mitropoulos et al. (67) and
(CTX), one of the accumulated products is also an in-
Mitropoulos and Myant (68) showed that there was a
termediate in the 25-hydroxylase pathway. Taken
small conversion of cholesterol into 27-hydroxy-
together, however, the metabolic consequences of the
cholesterol, 3@hydroxy-5cholenoic acid, lithocholic
metabolic blocks clearly support the contention that
acid, and chenodeoxycholic acid in rat liver mito-
the 27-hydroxylase pathway is the more important.
chondria fortified with cytosol (Fig.5 ) . Lithocholic
Second, the 25-hydroxylase pathway generates
acid is, however, a less efficient precursor to cheno-
acetone as a cleavage product, whereas the 27-
deoxycholic acid and all available evidence suggests
hydroxylase pathway generates propionic acid. In con-
that a pathway involving lithocholic acid as inter-
trast to acetone, propionic acid is rapidly oxidized to
mediate is of little or no importance under normal
carbon dioxide. By studying the relative formation of
conditions. Such a pathway may, however, be of some
radioactive carbon dioxide and acetone from choles-
importance under cholestatic conditions, for example.
terol labeled with 14C in the terminal methyl groups,
Thus it has been shown that urine from infants with
Duane et al. (65, 66) showed in both rat and human
biliary atresia and patients with liver disease contains
that the 25-hydroxylase pathway is a minor one under
elevated concentrations of 3fbhydroxy-5-cholenoicacid
normal conditions (Fig.4).
(for a review, see ref. 1).
Among the many possible alternative pathways in
SPECIES DIFFERENCES AND ALTERNATIVE S U E which the 27-hydroxyl group is introduced at a stage
STRATES FOR THE 27-HYDROXYLASE PATHWAY prior to completion of the nuclear changes (for a
review see ref. l ) , there is at present experimental sup-
In the sequence of reactions shown in Fig. 1, all the port for the presence of three specific such pathways
nuclear changes precede the changes in the steroid in human liver.

462 Joumal of Lipid Research Volume 33, 1992

 Most of the intermediates from this compound to

HO&- cholic acid were identified in the incubation medium

of these cells and the amounts of the products were
shown to change in a characteristic manner by an al-
teration of conditions. According to these authors
(72), cholic acid would be formed from 7a,12a-
dihydroxy4cholesten-hne by the sequence of 27-
hydroxylation, oxidation, and degradation of the side-
chain and A-ring reduction. An alternative pathway to
cholic acid was also described including reduction of
the intermediate 7a,12adihydroxy-3-ox&holesten-
oic acid to form 3a,7a,12a-trihydroxy-5~holestanoic
acid (THCA) prior to side chain cleavage (72).
HO’
Whether or not bile acids are formed by the same
mechanism in normal hepatocytes as in the above
Fig. 5. Mitochondrial conversion of cholesterol into chenodeoxy-
cholic acid (67, 68).
hepatoblastoma cells is not known with certainty. The
fact that two of the described intermediates,
7a,l2a,27-trihydroxy4cholesten-hneand 7a,12adi-
hydroxy-&x&holestenoic acid, are present in
Pathway involving 7a,27dihydroxy4cholesten-3-one human blood supports the contention that the path-
as intermediate way is significant (73).

Downloaded from www.jlr.org by guest, on May 14, 2019

7a-Hydroxy4cholesten-3-one is an efficient s u b
strate for the mitochondrial 27-hydroxylase in both rat
Pathway involving 27-hydroxycholesterolas
and human liver (4, 11). The product of the mito-
intermediate
chondrial 27-hydroxylase, 7a,27dihydroxy-4-cholesten-
&ne’, is rapidly converted into both cholic acid and Human liver mitochondria are able to 27-hydroxy-
chenodeoxycholic acid in vivo in humans (69, 70). late cholesterol (4). Krisans et al. (74) have shown that
Good evidence that this pathway is of importance in 27-hydroxycholesterol can be further metabolized into
humans is our finding that the lack of the mito- 3~hydroxy-5-cholenicacid in rat liver peroxisomes.
chondrial 27-hydroxylase in patients with CTX (see The cofactor requirement was the same as that in the
below) leads to a substantial accumulation of 7a- peroxisomal conversion of THCA into cholic acid
hydroxy4cholesten-3ne and its metabolite, 7a,12a- (74). Anderson, Kok, and Javitt (75) reported that 27-
dihydroxy4-cholesten-&ne (71). It is not known with hydroxycholesterol is rapidly converted into both
certainty how 7a,27dihydroxy4cholesten-3-oneis fur- cholic and chenodeoxycholic acid in vivo in humans.
ther metabolized. One alternative would be comple- 27-Hydroxycholesterol and 3~hydroxy-5-cholenoic
tion of the sidechain cleavage prior to further nuclearacid intermediates in the above pathway are present in
changes. Axelsson, Mbrk, and Everson (72) recently relatively high concentrations in human blood (76-
described such a pathway in cultured HepG2 cells. 78). We have found that the concentration of 27-
They isolated and identified all intermediates from 7a-hydroxycholesterol is considerably lower in the
hydroxy-4-cholesten-3-one to chenodeoxycholic acid circulation of rats and rabbits than in humans (I.
from the incubation medium. In this pathway side- Bjorkhem, unpublished observation) possibly indicat-
chain degradation was completed prior to 5breduc- ing that pathways involving 27-hydroxycholesterol are
tion of the A-ring. less important in the former species.
Axelsson and Sjbvall (73) and Axelsson, Mijrk, and
Sjbvall (76) have denoted the pathway to chenode-
Pathway involving 7a,12a,27-trihydroxyP oxycholic acid starting with 27-hydroxylation of choles-
cholesten-3-one as intermediate terol and proceeding via Csracids the “acid” pathway.
The accumulation of 7a,l2a-dihydroxy4-cholesten- The one starting with 7a-hydroxylation of cholesterol
&ne in patients with a lack of the mitochondrial 27- was called the “neutral” pathway. Since they found that
hydroxylase indicates that this compound may also be the levels of the circulating intermediates in the
an important substrate for the 27-hydroxylase in vivo neutral pathway were increased after cholestyramine
(71). In accordance with this, a pathway starting with treatment, whereas the intermediates in the acid path-
sidechain oxidation of 7a,12a-dihydroxy-4-cholesten- way were unaffected, they predicted that the two path-
&ne was very recently described in HepG2 cells (72). ways were regulated separately. They also studied the

Bjkkhem Mechanism of degradation of steroid side chain m formation of bile acids 463
behavior of 7a-hydroxy-3-ox&holestenoic acid in
the human circulation under different conditions
(73). This compound may be formed in both the
“neutral” and the “acid” pathway. In accordance with
this, the level of this acid was increased in the circula-
tion after cholestyramine treatment but not to the
same degree as 7a-hydroxy-4cholesten-3-one (73).
Further support for the presence of a pathway start-
ing with 27-hydroxylation comes from work by Princen
et a1 (79) and from collaborative work of H. Dahlback,
1. Bjorkhem, and H. Princen (unpublished observa-
tion). It was found that cyclosporin selectively inhibits
27-hydroxylation of cholesterol but not 27-hydroxyla-
tion of other 7a-hydroxylated bile acid intermediates
in liver mitochondria and in a reconstituted purified
cytochrome P450 system. Cholesterol 7a-hydroxylase
was not affected by the inhibitor. Hepatocytes from rat
and human hepatoblastoma cells (HepG2 cells)
responded with decreased bile acid formation when
exposed to cyclosporin. Such a decrease would be ex-
pected if part of the bile acids is formed in a pathway
involving 27-hydroxylation as a first and rate-limiting
step. It may be mentioned that cyclosporin has a hy-
percholesterolemic effect and it may be speculated
that this may be due in part to a blocking of the “acid”
pathway for bile acid biosynthesis.
If 27-hydroxycholesterol is an intermediate in a
pathway to cholic and chenodeoxycholic acid, the 7a-
hydroxyl group may be introduced at different stages.
It was recently reported (80) that 27-hydroxy-
cholesterol is efficiently 7a-hydroxylated in pig liver
mitochondria. In preliminary experiments we found
little or no 7a-hydroxylase activity toward 27-hydroxy-
cholesterol in human liver mitochondria (I. Bjorkhem,
E. Reihnkr, and K. Einarsson, unpublished observa-
tions). Significant such activity was found in the
microsomal fraction of a homogenate from human
liver. Another group (J. Shoda, A. Toll, M. Axelsson, F.
Pieper, K. Wikvall, and J. Sjovall, unpublished observa-
tions) found significant 7a-hydroxylase activity in both
the microsomal and mitochondrial fractions. Whether
or not the microsomal 7a-hydroxylase active towards
27-hydroxycholesterol is the same as that active
towards cholesterol is not known.
The relative importance of the pathway including
27-hydroxycholestero1 as an intermediate is not known
and one can only speculate about this. According to
the work by Axelson and Sjovall (73) and Princen et
al. (79), this pathway may be responsible for up to
50% of all bile acids formed in human liver under
basal conditions. In a situation where bile acid biosyn-
thesis is up-regulated, e.g., by a bile fistula or by
cholestyramine treatment, the contribution of the
“acid” pathway may be considerably less (73).
464 Journal of Lipid Research Volume 33, 1992
INBORN ERRORS IN SIDE-CHAIN CLEAVAGE
There are two known types of inborn metabolic er-
rors that affect cholesterol sidechain cleavage,
cerebrotendinous xanthomatosis (CTX) and various
peroxisomal disorders. In CTX there is a specific
defect in one of the enzymes. In the peroxisomal dis-
orders there is a lack of intact peroxisomes and thus
also a lack of the peroxisomal enzymes necessary for p
oxidation of cholestanoic acids.
Cerebrotendinousxanthomatosis
This rare, inherited lipid storage disease is charac-
terized by xanthomas, progressive neurological dys-
function, cataracts, and the development of
xanthomatous lesions in the brain and lung (for a
general review, see ref. 81). In contrast to other dis-
eases with tendon xanthomatosis, plasma cholesterol
levels are remarkably low. Large deposits of cholesterol
and cholestanol are present in most tissues. In 1974,
Setoguchi et al. (82) made the key discovery that CTX
Downloaded from www.jlr.org by guest, on May 14, 2019
patients have a defect in bile acid biosynthesis, with in-
complete oxidation of the C27-steroid sidechain, lead-
ing to excretion of large amounts of Cn~bilealcohols
in bile, feces, and urine. The formation of bile acids,
in particular chenodeoxycholic acid, was reduced.
There has been a controversy in the past between
our laboratory and that of Salen et al. concerning the
location of this defect (for a general review, see ref.
81). It is however, now established that the basic meta-
bolic defect is located to the mitochondrial 27-
hydroxylase. We have shown, for example, that
fibroblasts from CTX patients do not express this en-
zyme activity and that fibroblasts from heterozygotes
have an activity about 50% of normal (7). Very recent-
ly Cali et al. (83) showed that the 27-hydroxylase gene
in patients with CTX contained two cys-arg muta-
tions. When genes with these mutations were ex-
pressed in COScells, the expressed enzyme was found
to be inactive.
The lack of the mitochondrial 27-hydroxylase leads
to extensive accumulation of a number of substrates
for the enzyme, such as 7a-hydroxy-4-cholesten-3-one
and 5~cholestane-3a,7a,12a-triol. The accumulation
of 7c~-hydroxy-4-cholesten-30nemay be due in part to
the up-regulated cholesterol 7a-hydroxylase (see
below) which leads to increased levels of both 7a-
hydroxycholesterol and 7a-hydroxy-4-cholesten-3-one
in the liver and the circulation (81, 84, 85). The u p
regulation of the cholesterol 7a-hydroxylase is due to
the low formation of primary bile acids, in particular
chenodeoxycholic acid. Among the primary bile acids,
chenodeoxycholic acid seems to be the most efficient
suppressor of the human cholesterol 7&hydroxylase
(86). Since bile acids are involved also in the regula-
tion of the HMG-CoA reductase (1, 87, 88), cholester-
ol synthesis is also increased in CTX. As a
consequence, LDL turnover is increased (89). The
possibility has been discussed that the increased cho-
lesterol synthesis and LDL-turnover in CTX may be in
part a direct effect of the lack of 27-hydroxycholesterol
or its metabolites (30, 73). Treatment of patients with
chenodeoxycholic acid seems to normalize cholesterol
synthesis, however (90).
Because 7a-hydroxyPcholesten-%ne cannot be
metabolized by the usual pathway to bile acids, more
unusual pathways are used (Fig. 6 ) . We have shown
that one product of 7a-hydroxy-kholesten-%ne is
cholestanol (81). This compound is formed by
dehydration of 7a-hydroxy~holesten-3-oneto 46
cholestadien-%ne by a specific enzyme in liver
microsomes (91). The latter steroid is rapidly con-
verted into cholestanol (92). It appears likely that at
least a substantial part of the cholestanol accumulating
over the decades in the brain and tendons of patients
with CTX is formed from accumulated bile acid inter-
mediates.
Because the accumulated 5bholestane-3a,7a,12a-
triol cannot be metabolized by the usual pathway to
bile acids, the 25-hydroxylase pathway may be used
(Fig. 7). Since the enzymes involved in this pathway
appear to have a limited capacity, some of the inter-
mediates in this pathway accumulate and are excreted
in bile and feces. Thus patients with CTX excrete high
amounts of 5~holestane-3a,7a,l2a,Z5-tetrol(81, 82).
cho 1 e s t e r o 1 -&
HO
n ,
‘ ”OH
cholestanol
Fig. 6. Metabolic consequence of the accumulation of 7~-hydroxy4cholesten-3-onein patients with CTX.
Mechanism of formation of excess cholestanol (81).
Bjtirkhem Mechanism of degradation of steroid side chain in formation of bile acids
Also other metabolites of 5fkholestane-3a,7a, 1201-
triol, such as 5fkholestane-3a,l2a,12~~,24a-25pentol
and 5~holestane-3a,7a,l2a,23-25-pentol,are excret-
ed. In total, gram amounts of all the above bile alco-
hols are excreted daily in bile and feces. Treatment
with chenodeoxycholic acid suppresses the upregu-
lated cholesterol 7a-hydroxylase and results in a drasti-
cally reduced formation of bile alcohols and
cholestanol (90).
Peroxisomal disorders
The most serious of the peroxisomal diseases affect-
ing bile acid synthesis is Zellweger’s disease (for a
general review, see ref. 93). This rare congenital disor-
der is characterized by multiple craniofacial abnor-
malities, generalized hypotonia, central nervous system
abnormalities, hepatomegaly, and renal cortical cysts.
Most often the afflicted infants die within 6 months.
Other peroxisomal disorders also affecting bile acid
biosynthesis are infantile Refsum’s disease, and neona-
tal adrenoleukodystrophia (93). Patients with Zell-
Downloaded from www.jlr.org by guest, on May 14, 2019
weger’s disease lack intact peroxisomes and most of
the peroxisomal enzymes are reduced or absent. Con-
sequently these patients have several severe metabolic
disturbances (reduced capacity to oxidize fatty acids,
prostaglandins, phytanic acid, reduced synthesis of
plasmalogens) (93).
In accordance with our finding that peroxisomes
contain the enzymes involved in bxidation of the
CoA derivative of trihydroxycoprostanoic acid, patients
with Zellweger’s disease, infantile Refsum’s disease,
chenodeoxy
cholic acid
465
 cholesterol -#-
- &e
HO HO'
cilohT:&

HO'
acid

HO'
&OH

"OH HO'
& 8 "OH
OH

HO' "OH
OH .....,......,.

Fig. 7. Metabolic consequence of the accumulation of 5~holestane-Sa,7a,l2a-uiolin patients with CTX.

Mechanism of formation of C ~ b i l ealcohols and utilization of the 25-hydroxylase pathway for formation of
cholic acid.

Downloaded from www.jlr.org by guest, on May 14, 2019

and neonatal adrenoleukodystrophia accumulate
THCA and a number of metabolites of this compound
(for general reviews, see refs. 93 and 94). The most im-
portant metabolites that accumulate in the above dis-
eases are summarized in Fig. 8. One interesting and
quantitatively important metabolite of THCA acid is a
dicarboxylic Cwacid (94). The mechanism behind the
formation of this chain-elongated derivative is not
known. It should be emphasized that the metabolic
block is not complete since patients with above
peroxisomal disorders also form some cholic acid. In
collaborative studies together with Pedersen and Kase
we have shown, however, that the cholic acid and
chenodeoxycholic acid pool sizes are drastically
reduced (95, 96). After administration to Zellweger
patients, different labeled intermediates in bile acid
biosynthesis, carrying an intact Czrsteroid sidechain,
were found to be rapidly converted into THCA but
then only very slowly to cholic acid (95, 96). A liver
biopsy from a patient with Zellweger's disease had no
significant ability to convert THCA into cholic acid
(96). There was, however, some conversion of 301,7a-
dihydroxy-5f3-cholestanoicacid into chenodeoxycholic
acid. The latter finding may explain why there is little
or no accumulation of 301,7~tdihydroxy-5~holes-
tanoic acid in this disease. Another explanation is that
301,7~xdihydroxy-5~holestanoic acid is efficiently 1201-
hydroxylated by the human liver (97). Yet, another ex-
planation was presented by Axelson et al. (72) who
found an accumulation of 7c~hydroxy-3-ox&holes

&COOH fl COOHCOOH

HO' "OH HO' "OH HO' "OH

OH
Fig. 8. Metabolic consequence of the accumulation of THCA in patients with the Zellweger disease. Only
the quantitatively most dominating metabolites of THCA are shown.

466 Journal of Jipid Research Volume 33, 1992


Sitosterol Stigmasterol Campesterol
Fig. 9. Structures of the most important phytosterols.
tenoic acid and its hydrolyzed product, 3-0x0-4,6-
cholestadienoic acid, in blood from a patient with this
disease. This finding suggests that 7a-hydroxy-30x04
cholestenoic acid could be the substrate for the
peroxisomal sidechain degradation instead of 3a,'7a-
dihydroxy-5@holestanoic acid.
In any case, it is evident that the accumulation of
THCA acid and its metabolites in the peroxisomal dis-
eases reflects the importance of the peroxisomes for
the last step in the biosynthesis of cholic acid from
cholesterol.
MECHANISM OF DEGRADATION OF THE
STEROID SIDE CHAIN OF PLANT STEROIDS
The structure of the three most common plant
sterols is shown in Fig. 9. These plant sterols are nor-
mal constituents of the human diet. They are ab-
sorbed to a very limited extent from the intestine,
circulate in plasma in low concentrations, and are ex-
creted in bile (for a review, see ref. 81).
In theory, the presence of an ethyl group at C-24
should prevent or at least obstruct conversion of the
plant sterols to bile acids. The possibility must be con-
sidered that sitosterol is first dealkylated to cholesterol
and then converted into bile acids. Such a dealkylation
occurs in some worms and crabs (98, 99) but attempts
to demonstrate such a pathway in mammals have
failed hitherto. In accordance with this, attempts to
demonstrate conversion of sitosterol into normal C24-
bile acids in rats (100) and monkeys (101) have failed.
In an early study, Salen, Ahrens, and Grundy (102)
reported an efficient formation of cholic and
chenodeoxycholic acid from intravenously admin-
istered [22,2S3H]sitosterol in humans. In a reinves-
tigation with [4I4C]sitosterol we could not find any
significant conversion into labeled (&-bile acids in two
healthy subjects (103). In order to bypass the rate-
limiting step, the metabolic fate of "H-labeled '7a-
hydroxysitosterol was also studied. In this case there
was a significant conversion into acid products in bile.
Although part of the labeled products had the chro-
matographic properties of cholic and chenode-
oxycholic acid, further analysis showed that none of
Bjiirkhem Mechanism of degradation of steroid side chain m formation of bile acids
the products was identical to chenodeoxycholic acid
and only traces at the most could be identical to cholic
acid. The results suggested that healthy human sub-
jects, like other mammalian species studied, have little
or no capacity to convert sitosterol into the normal
Cpqbile acids.
Several previous studies have shown that sitosterol is
converted into polar compounds in the bile acid frac-
tion of rat bile (100, 104, 105). Very recently we were
able to identify most of these products (106, 107). The
major part of the I4C radioactivity recovered as bile
acids in bile after intravenous administration of [4
l4C1sitosterol to female Wistar rats was found to be
considerably more polar than cholic acid and only
trace amounts had chromatographic properties similar
to those of cholic acid and chenodeoxycholic acid. It
was shown that the polar products were di- and
trihydroxylated Czl-bile acids.
Using mass spectrometry, NMR, stereospecific
dehydrogenases and reagents, the major trihydroxy-
lated Czl-bile acids were identified as 5ppregnan-
Downloaded from www.jlr.org by guest, on May 14, 2019
3~,11f3,15~trio1-21-oic acid, 5ppregnan-Sa,l lp,15a-21-
oic acid, and 5Ppregnan-Sa,l lf3,16-triol-21-oic acid.
The major dihydroxylated &-bile acid was identified
by the same means as 5a-pregnan-3a,l2~l-diol-21-oic
acid (Fig. 10) (see refs. 97, 98).
Considerably less Czl-bile acids were formed from
labeled sitosterol in male than in female Wistar rats.
The Czl-bile acids formed in male rats did not contain
a 15-hydroxyl group. Conversion of labeled sitosterol
into Czl-bile acids also occurred in adrenalectomized
and ovariectomized rats, indicating that endocrine tis-
sues were not involved. Experiments with isolated per-
fused liver gave direct evidence that the overall
conversion of sitosterol into C2l-bile acids occurs in
this organ. Attempts to demonstrate production of
Cnl-bile acids in isolated hepatocytes have failed
hitherto.
After intravenous injection of deuterium-labeled
sitosterol in bile fistula female Wistar rats, the isolated
Czl-bile acids were found to contain very little isotope,
no .Acoon no
AcaoH
no no OY-
Fig. 10. Structures of the major Czi-bile acids formed from
cholesterol and plant sterols in female Wistar ratS (97, 98).
467
indicating a high dilution with endogenous material.
In view of the small amounts of sitosterol in the diet, it
appeared likely that Czl-bile acids may also be formed
from a more abundant precursor, such as cholesterol.
It was subsequently shown, using a mixture of [4
‘‘C]cholesterol and [22-3H]cholesterol, that there was
an efficient conversion of the administered material
into the above C21-bile acids. The conversion was
surprisingly high, up to about 25%. It is thus evident
that a blocking group at C24 is not obligatory for
degradation of the steroid side-chain beyond the C24
stage. It was further shown that campesterol, carrying a
methyl group in 24position, was also converted into
the above C21-bile acid in female Wistar rats (106).
The mechanism for formation of the Czl-bile acids
can only be speculated about at the present stage of
knowledge. The most likely mechanism seems to be a
primary attack by a mixed function oxidase at C21, fol-
lowed by further oxidation to give a C2lcarboxylic
group. The latter compound may then be further
oxidized to give a C21-bile acid as final product (Fig.
11).At present, however, there is no direct evidence of
this hypothetical “21-hydroxylase pathway.” In prelimi-
nary experiments we synthesized labeled 21-
hydroxycholesterol and administered it to female
Wistar rats. Most of the polar products obtained were
not identical to the above Cpl-bile acids. It is thus likely
that some metabolic step(s) in the steroid nucleus
precedes the 21-hydroxylation. It is tempting to sug-
gest that the sex-specific 15-hydroxylation may
facilitate the formation of C2l-bile acids.
The high capacity of the female rat liver to convert
cholesterol into compounds considerably more polar
than the usual bile acids may be of some regulatory
importance in the overall cholesterol balance. The
role of Czl-bile acids in other mammalian species is
uncertain, however. In humans there seems to be little
or no formation of such bile acids (E. Lund, K. Muri-
Boberg, and I. Bjorkhem, unpublished observation).
---
0-ox
HOOC
R
Fii. 11. Suggested mechanism for formation of Czi-bile acids from
phytosterols and cholesterol in female Wistar rats (97, 98).
468 Journal of Lipid Research Volume 33, 1992
CONCLUDING REMARKS
Three different mechanisms are known for the
degradation of the cholesterol side chain.
The major mechanism, the “27-hydroxylase path-
way,” involves a primary attack by a mitochondrial
cytochrome P-450 system followed by peroxisomal
oxidation. Metabolic blocks in this pathway lead to ac-
cumulation of intermediates, with severe metabolic
consequences.
The “25-hydroxylase pathway” involves a primary at-
tack by a microsomal 25hydroxylase. This pathway has
a low capacity and seems to be of importance only
when the 27-hydroxylase pathway is blocked.
In addition to the above systems, a complementary
system has been evolved in at least one mammalian
species that is able to degrade both cholesterol and
plant sterols into highly polar Cgl-bile acids. It is prob-
able that this degradation is initiated by attack of a 21-
hydroxylase. The relative importance of the
hypothetical “21-hydroxylase pathway” in different
Downloaded from www.jlr.org by guest, on May 14, 2019
mammalian species is not known. I
The author is most grateful to the following colleagues for
reading the manuscript, for consuuctive criticism, and for
information concerning unpublished studies: Dr. Magnus
Axelson, Dr. Kyuchiro Okuda, Dr. Jan I. Pedersen, Dr. Hans
Princen, Dr. Jan Sjbvall, and Dr. Kjell Wikvall. The author’s
own work, referred to in the review, has been supported by a
grant from the Swedish Medical Research Council.
Manusnipt received 21 Noumber I991
REFERENCES
1. BjBrkhem, I. 1985. Mechanism of bile acid biosynthesis
in mammalian liver (Review). In Comprehensive
Biochemistry. Vol. 12. H. Danielsson and J. Sjbvall,
editors. Elsevier Publishing Co., Amsterdam. 231-278.
2. Popjak, G., J. Edmond, F. A. L. Anet, and N. R. Eaton,
Jr. 1977. Carbon-13 NMR studies on cholesterol syn-
thesized from 13C-mevalonates.J. Am. Chem. SOC. 9 9
931-935.
3. Gustafsson, J., and S. Sjbstedt. 1978. On the stereo-
specificity of microsomal “26”-hydroxylationin bile acid
biosynthesis. J. Biol. C h . 253: 199-201.
4. BjBrkhem, I., J. Gustafsson, G. Johansson, and B.
Persson. 1975. Biosynthesis of bile acids in man:
hydroxylation of the C27-steroid side chain. J. Clin. In-
vest. 55: 478-486.
5. Hanson, R. F., P. Szczepanik, and G. C. Williams. 1980.
Stereochemistry of the side chain oxidation of 5p
cholestaneSa,7a,lfa-triolin man. J. Biol Chem. 255:
1483-1485.
6. Bjbrkhem, I., H. Danielsson, and J. Gustafsson. 1973.
On the effect of thyroid hormone on 26hydroxylation
of Cgpsteroids in rat liver: FEBS Lett. 31: 20-22.
7. Skrede, S., I. Bjcrkhem, E. A. Kvittingen, M. S. Buch-
mann, S. 0. Lie, C. East, and S. Grundy. 1986.
Demonstration of 26hydroxylation of Cq7-steroids in
human skin fibroblasts and deficiency of this activity in
CTX. J. Clin. Invest. 78: 729-735.
8. Pedersen, J. I., H. Oftebro, and I. Bjdrkhem. 1989.
Reconstitution of Cp7-steroid 26hydroxylase activity
from bovine brain mitochondria. Biochem. Znt. 1 8 615-
622.
9. Postlind, H., and K. Wikvall. 1989. Evidence for the for-
mation of 26hydroxycholesterol by cytochrome P450
in pig kidney mitochondria. Biochem. Biophys. Res. Com-
mun. 159: 1135-1140.
10. Anderson, S., D. L. Davis, H. Dahlbick, H. Jdrnvall,
and D. W. Russell. 1989. Cloning, structure, and expres-
sion of the mitochondrial cytochrome P-450 sterol 2 6
hydroxylase, a bile acid biosynthetic enzyme. J. Biol.
Chem. 264 8222-8229.
11. Thompson, S. L., and S. K. Krisans. 1985. Evidence for
peroxisomal hydroxylase activity in rat liver. Biochem.
Biophys. Res. Commun. 1 3 0 708-716.
12. Bjdrkhem, I., and J. Gustafsson. 1973. @Hydroxylation
of steroid sidechain in biosynthesis of bile acids. Eur. J.
Biochem. 36: 201-212.
13. Bjdrkhem, I., and J. Gustafsson. 1974. Mitochondrial w
hydroxylation of the cholesterol side chain. J. Biol.
Chem. 249: 2528-2535.
14. Taniguchi, S., N. Hoshita, and K. Okuda. 1973. En-
zymatic characteristics of CO-sensitive 26hydroxylase
system for 5~holestane-3a,7a,12a-triolin rat liver
mitochondria and its intramitochondrial localization.
Eur.J.’Biochem. 40: 607-617.
15. Okuda, IC,P. Weber, and V. Ullrich. 1977. Photochemi-
cal action spectrum of the CO-inhibited 5fiholestane
3a,7a, 12a-triol 26hydroxylase system. Biochem. Biophys.
Res. Commun. 7 4 1071-1075.
16. Gustafsson, J. 1976. On the heterogeneity of the
mitochondrial Cg7-steroid 26hydroxylation system. J.
Lipid Res. 17: 366-372.
17. Pedersen, J., H. Oftebro, and T. Vinngird. 1977. Isola-
tion from bovine liver mitochondria of a soluble fer-
redoxin active in a reconstituted steroid hydroxylating
system. Biochem. Biophys. Res. Commun. 76: 666-673.
18. Sato, R., Y. Atsuta, Y. Imai, S. Taniguchi, and K. Okuda.
1977. Hepatic mitochondrial cytochrome P-450: isola-
tion and functional characterization. R o c . Natl. Acad.
Sn’. USA 74: 5477-5481.
19. Pedersen, J. I., I. Bjdrkhem, and J. Gustafsson. 1979. 2 6
Hydroxylation of C27-steroids by soluble liver mitochon-
drial cytochrome P-450. J. Biol. Chem. 254 64646469.
20. Wikvall, K. 1984. Hydroxylations in biosynthesis of bile
acids: isolation of a cytochrome P-450 from rabbit liver
mitochondria catalyzing 26hydroxylation of C27-
steroids. J. Biol. Chem. 259: 3800-3804.
21. DahlbLk, K., and K. Wikvall. 1988. 25Hydroxylation of
vitamin D3 by a cytochrome PA50 from rabbit liver
mitochondria. Biochem.J. 252 207-213.
22. Okuda, IC, 0. Masumoto, and Y. Ohyama. 1988.
Purification and characterization of 5pcholestane-
31~,7a,lBa-triol27-hydroxylase from female rat liver
mitoch0ndria.J Biol. Chem. 263: 18138-18142.
23. Cali, J. J., and D. W. Russell. 1991. Characterization of
human sterol 27-hydroxylase: a mitochondrial
cytochrome P450 that catalyzes multiple oxidations in
bile acid biosynthesis.J. Biol. Chem. 266: 77747778.
24. Bergman, T., and H. Postlind. 1991. Characterization of
Bj&khem Mechanism of degradation of steroid side chain in formation of bile acids
mitochondrial cytochrome P450 from pig kidney and
liver catalyzing 26hydroxylation of 25hydroxyvitamin
D3 and C2’1-steroids. Biochem.J. 276: 427-432.
25. Gustafsson, J. 1978. Effect of biliary obstruction on 2 6
hydroxylation of Cpsteroids in bile acid synthesis. J.
Lipid Res. 19: 237-243.
26. Saarem, K., and J-I. Pedersen. 1987. Sex differences in
the hydroxylation of cholecalciferol and of 5bholes-
tane-3a,7a,l2a-triol in rat liver. Biochem.J. 247: 73-78.
27. Dahlbick, H. 1989. Thesis. Uppsala University.
Lindbergs Grafiska H. B.
28. Usui, E., M. Noshiro, and K. Okuda. 1990. Molecular
cloning of cDNA for vitamin D3 25hydroxylase from rat
liver mitochondria. FEBS Lett. 262 135-138.
29. Raza, H., and N. G. Avadhani. 1988. Hepatic
mitochondrial cytochrome P-450 system: purification
and characterization of two distinct forms of mito-
chondrial cytochrome P450 from knaphthoflavone-in-
duced rat liver. J. Biol. Chem. 263: 9533-9541.
30. Javitt, N. B. 1990. 26HydroxycholesteroI: synthesis,
metabolism, and biologic activities. J. Lipid Res. 31:
1527-1533.
31. Esterman, A. L., H. Baum, N. B. Javitt, and G. J. Dar-
lington. 1983. 26Hydroxycholestero1: regulation of
hydroxymethylglutaryl-CoA reductase activity in
Downloaded from www.jlr.org by guest, on May 14, 2019
Chinese hamster ovary cell culture. J Lipid Res. 2 4
1304-1 309.
32. Pedersen, J. I., I. Holmberg, and I. Bjdrkhem. 1979.
Reconstitution of vitamin D3 25-hydroxylase activity
with a cytochrome P450 preparation from rat liver
mitochondria. FEBS Lett. 9 8 394-398.
33. Bjdrkhem, I., I. Holmberg, H. Oftebro, and J. I. Peder-
sen. 1980. Properties of a reconstituted vitamin D3 2 5
hydroxylase from rat liver mitochondria. J. Biol. Chem.
255: 52445249.
34. Dahlbick, H. 1988. Characterization of the liver
mitochondrial cytochrome P-450 catalyzing the 26-
hydroxylation of 5bholestane-3a,7a,l2a-triol.Biochem.
Biophys. Res. C?mmun. 157: 30-36.
35. Ohyama, Y., 0. Masumoto, E. Usui, and K. Okuda.
1991. Multi-functional property of rat liver mitochon-
drial cytochrome P450. J. Bdochem. (Tokychem (Tokyo)
109: 389-393.
36. Okuda, K., and H. Danielsson. 1965. Synthesis and me-
tabolism of 5~holestane-Sa,7c~,12a-trio1-26al. Acta
Chem. Scand. 19: 2160-2165.
37. Okuda, K., and N. Takigawa. 1969. The dehydrogena-
tion of 5&cholestane-3a,7a,12a,26tetrolby rat liver.
Biochlm. Biophys. Acta. 176: 873-879.
38. Okuda, IC, and N. Takigawa. 1968. Separation of 5p-
cholestane-3a,7a,l2a,26-tetrol oxidoreductase, ethanol
oxidoreductase, 5~holestane-3a,7a,12a-triol-26al oxi-
doreductase and acetaldehyde oxidoreductase from rat
liver. Biochem. Biophys. Res. Commun. 3 3 788-793.
39. Okuda, K., and N. Takigawa. 1970. Rat liver 5bholes-
tane-3a,7a, 12a,26tetrol dehydrogenase as a liver alco-
hol dehydrogenase. Biochim. Biophys. Acta. 222:
141-148.
40. Waller, G., H. Theorell, and J. Sjdvall. 1965. Liver al-
cohol dehydrogenase as a 3j3-hydroxy-5f3-cholanic acid
dehydrogenase. Arch. Biochem. @i%hys. 111: 671-684.
41. Bjdrkhem, I., H. JBrnwall, and A. Akesson. 1974. Oxida-
tion of *hydroxylated fatty acids and steroids by S S
isoenzyme of liver alcohol dehydrogenase. Biochem.
Biophys. Res. Commun. 57: 870-875.
42. Dahlbick, H., H. Danielsson, M. Gustafsson, J. Sjdvall,
469
 and R Wikvall. 1988. Conversion of 5P-cholestanoic
acid by rabbit liver mitochondria. Biochem. Biophys. Res.
Commun. 153: 267-274.
43. Dahlbkk, H., and 1. Holmberg. 1990. Oxidation of 5 p
cholestane-3a,7a,l2a-triol into 3a,7a,l2a-trihydroxy-
5bholestanoic acid by cytochrome P-45026 from rabbit
liver mitochondria. Biochem. Biophys. Res. Commun. 167:
- ...391-395. . .___._
44. SjGvall, J., S. H. G. Anderson, and C. S. Lieber. 1987
Bile acids in deer mice lacking liver alcohol dehy-
drogenase. Biochim. Biophys. Acta. 836: 8-1 3.
45. Masui, T., and E. Staple. 1966. The formation of bile
acids from cholesterol. J. Biol. Chem. 241: 3889-3893.
46. Gustafsson, J. 1975. Biosynthesis of cholic acid in rat
liver. 24Hydroxylation of 3a,7a,l2a-trihydroxy-5f3-cho- 63.
lestanoic acid. J. Biol. Chem. 250: 8243-8247.
47. Masui, T., and E. Staple. 1965. The formation of cholic
acid from 3a,7a,12a,24~tetrahydroxycoprostanic acid
by rat liver. Biochim. Biophys. Acta. 104 305-307.
48. Gustafsson, J. 1980. Biosynthesis of cholic acid in rat
liver: formation of cholic acid from 3a,7a,12a-
trihydroxy- and 3a,7a,12a,24tetrahydroxy-5pcholes-
tanoic acids. Lipids. 1 5 113-121.
49. Pedersen, J. I., and J. Gustafsson. 1980. Conversion of
3a,7a,l2a-trihydroxy-5~holestanoicacid into cholic
acid by rat liver peroxisomes. EEBS Lett. 121: 345-348.
50. Kase, F., I. Bjijrkhem, and J. I. Pedersen. 1983. Forma-
tion of cholic acid from 3a,7a,12a-trihydroxy-5fl-choles-
tanoic acid by rat liver peroxisomes. J. Lipid Res. 2 4
1560-1 567.
51. Prydz, IC, B. F. Kase, I. Bjbrkhem, and J. I. Pedersen.
1986. Formation of chenodeoxycholic acid from 3a,7a-
dihydroxy-5pcholestanoicacid by rat liver peroxisomes.
J. Lipid Res. 27: 622-628.
52. Bjijrkhem, I., B. F. Kase, and J. I. Pedersen. 1984.
Mechanism of peroxisomal 24hydroxylation of
3a,7a,l2a-trihydroxy-5~holestanoic acid in rat liver.
Biochim. Biophys. Acta. 796: 142-145.
53. Ostlund-Farrants, A. R, I. Bjbrkhem, and J-I. Pedersen.
1989. Identification of 3a,7a,l2a-trihydroxy-5~holest-
24-enoic acid as an intermediate in the peroxisomal
conversion of 3t~,7a,l2a-trihydroxy-5~holestanoic
acid to cholic acid. Biochim. Biophys. Acta. 1002: 198-
202.
54. Prydz, R,B. F. Kase, I. Bjbrkhem, and J. I. Pedersen.
1988. Subcellular localization of 3a,7adihydroxy- and
3a,7a,12a-trihydroxy-5~cholestanoyl-CoA-ligase (s) in
rat liver. J. Lipid Res. 29: 997-1004.
55. Pedersen, J. I., E. Hvattum, T. Flatabo, and I. Bjbrkhem.
1988. Clofibrate does not induce peroxisomal
3a,7a,12a-trihydroxy-5Pcholestanoyl CoA oxidation in
rat liver. Evidence that this reaction is catalyzed by an
enzyme different from that of peroxisomal acyl-CoA
oxidation, Biochem. Znt. 17: 163-169.
56. Farrants-Ostlund, A. R , I. BjGrkhem, and J. I. Pedersen.
1990. Differential induction of peroxisomal oxidation
of palmitic acid and 3a,7a,12a-trihydroxy-5~holes-
tanoic acid in rat liver. Biochim. Biophys. Acta. 1046: 173-
177.
57. Schepers, L., P. P. van Veldhoven, M. Casteels, H. J. Eys-
sen, and G. P. Mannaerts. 1990. Presence of three acyl-
CoA oxidases in rat liver peroxis0mes.J. Biol. Chem. 265
5242-5246.
58. Gengenbacher, T., W. Gerok, U. Gise, W. W. Just, and
G. Kurz. 1991. Peroxisomal proteins involved in bile salt
biosynthesis. In Bile acids as Therapeutic Agents.
Kluwer Academic Publishers, Norwell, MA. 63-76.
470 lournal of finid Research Volume 33. 1992
.#
59. h e , B. F., K. Prydz, I. Bjbrkhem, and J. I. Pedersen.
1986. Conjugation of cholic acid with taurine and
glycine by rat liver peroxisomes. Biochem. Biophys. Res.
Commun. 138: 167-1 73.
60. Kase, B. F., and I. Bjbrkhem. 1989. Peroxisomal bile
acid-CoAamino acid N-acyltransferase in rat liver. J.
Biol. Chem. 264: 9220-9223.
61. Johnson, M. R., S. Barnes, J. B. Kwakye, and R. B.
Diasio. 1991. Purification and characterization of bile
acid-CoA:amino acid N-acyltransferase from human
liver. J. Biol. Chem. 266: 10227-10233.
62. Cronholm, T., and G. Johansson. 1970. Oxidation of
5pcholestane-3a,7a,l2~~triol by rat liver microsomes.
Eur. J. Biochem. 16: 373-381.
Shefer, S., F. W. Cheng, B. Dayal, S. Hauser, G. S. Tint,
G. Salen, and E. H. Mosbach. 1976. A 25-hydroxylase
pathway of cholic acid in man and rat. J. Clin. Invest. 57:
897-903.
64. Salen, G., S. Shefer, F. W. Cheng, B. Dayal, A. K. Batta,
and G. S. Tint. 1979. Cholic acid biosynthesis: the en-
zymatic defect in cerebrotendinous xanthomatosis. J.
Clin. Invest. 6 3 38-44.
65. Duane, W., I. Bjbrkhem, J. N. Hamilton, and S. M.
Mueller. 1988. Quantitative importance of the 25-
hydroxylase pathway for bile acid biosynthesis in the
Downloaded from www.jlr.org by guest, on May 14, 2019
rat. Hepatoha. 8 613-618.
66. Duane, W. C., P. A. Pooler, and J. N. Hamilton. 1981.
Bile acid synthesis in man. In vivo activity of the 25-
hydroxylase pathway. J. Clin. Invest. 8 2 82-85.
67. Mitropoulos, K. A., M. D. Avery, N. B. Myant, and G. F.
Gibbons. 1972. The formation of cholest-5ene-3P,26
diol as an intermediate in the conversion of cholesterol
into bile acids by liver mitochondria. Biochem. J. 130
363-371.
68. Mitropoulos, K. A., and N. B. Myant. 1967. The forma-
tion of lithocholic acid, chenodeoxycholic acid and p
muricholic acid from cholesterol incubated with rat
liver mitochondria. Biochem.J. 103 472-479.
69. Swell, L., J. Gustafsson, C. C. Schwartz, L. G. Halloran,
H. Danielsson, and Z. R. Vlahcevic. 1980. An in vivo
evaluation of the quantitative significance of several
potential pathways to cholic and chenodeoxycholic
acids from cholesterol in man.J. Lipid Res. 21: 455-466.
70. Bjbrkhem, I., 0. Fausa, G. Hopen, H. Oftebro, J. I.
Pedersen, and S. Skrede. 1983. Role of the 2 6
hydroxylase in the biosynthesis of bile acids in the nor-
mal state and in CTX.J. Clin. Invest. 71: 142-148.
71. Bjbrkhem, I., H. Oftebro, S. Skrede, and J. I. Pedersen.
1981. Assay of intermediates in bile acid biosynthesis
using isotope dilution-mass spectrometry: hepatic
levels in the normal state and in cerebrotendinous xan-
thomatosis.J. Lipid Res. 2 2 191-200.
72. Axelson, M., B. Mbrk, and G. T. Everson. 1991. Bile
acid synthesis in cultured human hepatoblastoma cells.
J. Biol. Chem. 266: 17770-17777.
73. Axelson, M., and J. Sjbvall. 1990. Potential bile acid
precursors in plasma, possible indicators of biosynthetic
pathways to cholic and chenodeoxycholic acid in man.
J. Steroid Biochem. 3 6 631-640.
74. Krisans, S. IC, S. L. Thompson, L. A. Pena, E. Kok, and
N. B. Javitt. 1985. Bile acid synthesis in rat liver
peroxisomes: metabolism of 26-hydroxycholesterol to
3~hydroxy-5-cholenoicacid.J. Lipid Res. 2 6 1324-1332.
75. Anderson, IL E., E. Kok, and N. BrJavitt. 1972. Bile acid
synthesis in man: metabolism of 7a-hydroxycholesterol-
14C and 26hydro~ycholesterol-~H. J. Clin. Invest. 51:
112-1 17.
76. Axelson, M., B. MBrk, and J. SjBvall. 1991. Studies on
biosynthetic pathways to cholic and chenodeoxycholic
acids in humans. In Bile acids as Therapeutic Agents.
Falk Symposium no. 58. Kluwer Academic Publishing
Co., Norwell, MA. 53-62.
77. Axelson, M., B. MBrk, and J. SjGvall. 1988. Occurrence
of 3~hydroxy-5-cholestenoic acid, 3j3,7adihydroxy-5-
cholestenoic acid and 7a-hydroxy-3-ox&holestenoic
acid as normal constituents in human blood. J. Lipid
Res. 29: 629-641.
78. Javitt, N. B., E. Kok, B. Cohen, and S. Bursteon. 1982.
Cerebrotendinous xanthomatosis: reduced serum 26-
hydroxycholesterol. J. Lipid Res. 23: 627-630.
79. Princen, H. M. G., P. Meijer, B. G. Wolthers, R. J. Vonk,
and F. Kuipers. 1991. Cyclosporin A blocks bile acid
synthesis in cultured hepatocytes by inhibition of
chenodeoxycholic acid and Pmuricholic acid synthesis.
B i o c h a . J. 2 7 5 501-505.
80. Axelson, M., J. Skoda, J. SjBvall, A. Toll, and K. Wikvall.
1992. Cholesterol is converted to 7a-hydroxy-33x04
cholestenoic acid in liver mitochondria-evidence for a
mitochondrial sterol 7a-hyroxylase. J. Biol. Chem. 267:
1701-1 704.
81. BjBrkhem, I., and S. Skrede. 1989. Cerebrotendinous
xanthomatosis and phytosterolemia. Chapter 51. In The
Metabolic Basis of Inherited Disease. C. R. Scriver, A. L.
Beaudet, W. S. Sly, and D. Valle, editors. McGraw-Hill
Book, Co., New York. 1283-1302.
82. Setoguchi, T., G. Salen, G. S. Tint, and E. H. Mosbach.
1974. A biochemical abnormality in cerebrotendinous
xanthomatosis. Impairment of bile acid biosynthesis a s
sociated with incomplete degradation of the cholesterol
side chain.J. Clin. Invest. 531: 1393-1401.
83. Cali, J. J., GL. Hsieh, U. Francke, and D. W. Russel.
1991. Mutations in the bile acid biosynthetic enzyme
sterol 27-hydroxylase underlie cerebrotendinous xan-
thomatosis. J. Biol. Chem. 266 7779-7783.
84. Bjdrkhem, I., E. Reihntr, B. A. Angelin, S. Ewerth, J-E.
Akerlund, and K. Einarsson. 1987. On the possible use
of the serum level of 7a-hydroxycholestero1 as a marker
for increased activity of the cholesterol 7a-hydroxylase
in humans. J. Lipid Res. 28: 889-894.
85. Axelson, M., A. Aly, and J. SjBvall. 1988. Levels of 7a-
hydroxykholesten-30ne in plasma reflect rates of bile
acid synthesis in man. FEBS Lett. 239: 324-328.
86. Reihnkr, E., I. BjBrkhem, B. Angelin, S. Ewerth, and K.
Einarsson. 1989. Bile acid synthesis in humans: regula-
tion of hepatic cholesterol 7a-hydroxylase activity.
Gastroenterology. 97: 1498-1505.
87. Akerlund, J. E., and I. Bjdrkhem, and U. Andersson.
1990. Studies on the regulation of cholesterol 7a-
hydroxylase and HMGCoA reductase in rat liver: ef-
fects of lymphatic drainage and ligation of the lymph
duct. J. Lipid Res. 31: 2159-2166.
88. Duckworth, P. F., Z. R. Vlahcevic, E. J. Studer, E. C.
Gurley, D. M. Heuman, Z. H. Beg, and P. B. Hylemon.
1991. Effect of hydrophobic bile acids on 3-hydroxy-3-
methylglutaryl-CoA reductase activity and mRNA levels
in the rat. J. Biol. Chem. 266: 9413-9418.
89. Ballantyne, C. M., G. L. Vega, C. East, G. Richards, and
S. M. Grundy. 1987. Lowdensity lipoprotein metabo-
lism in cerebrotendinous xanthomatosis. Metabolism. 36:
270-276.
90. Salen, G., T. W. Meriwether, and G. Nicolau. 1975.
Chenodeoxycholic acid inhibits increased cholesterol
and cholestanol synthesis in patients with cerebroten-
dinous xanthomatosis. Biochem. Med. 14: 57-74.
B j i i r k h Mechanism of degradation of steroid side chain m formation of bile acids
91. Skrede, S., M. S. Buchmann, and I. Bjdrkhem. 1988.
Hepatic 7adehydroxylation of bile acid intermediates,
and its significance for the pathogenesis of cerebroten-
dinous xanthomatosis. J. Lipid Res. 29: 157-164.
92. Kandutsch, A. A. 1963. Metabolism of cholesta4,7dien-
%ne and cholesta4,6dien-3-one by mouse liver micro-
somes. J. Lipid Res. 4 179-187.
93. Lazarow, P. B., and H. Moser. 1989. Disorders of
peroxisome biogenesis. In The Metabolic Basis of In-
herited Disease. Chapter 57. C. R. Scriver, A. L.
Beaudet, W. S. Sly, and D. Valle, editors, McGraw-Hill
Book Co., New York. 1479-1509.
94. Pedersen, J. I., B. F. Kase, K. Prydz, and 1. Bjdrkhem.
1987. Role of peroxisomes in bile acid formation.
Preceedings of the X International Bile Acid Meeting,
Freiburg, June 1988. Kluwer Academic Publishers, Nor-
well, MA. 41-46.
95. Kase, B. F., J. 1. Pedersen, B. Strandvik, and I.
Bjdrkhem. 1985. In vivo and in vitro studies on forma-
tion of bile acids in patients with Zellweger syndrome.
Evidence that peroxisomes are of importance in the
normal biosynthesis of both cholic and cheno-
deoxycholic acid. J. Clin. Invest. 76: 2393-2402.
96. Kase, B. F., I. Bjdrkhem, P. Higi, and J. I. Pedersen.
1985. Defective peroxisomal cleavage of the Cz;r-steroid
Downloaded from www.jlr.org by guest, on May 14, 2019
sidechain in cerebrohepatorenal syndrome of Zell-
weger. J. Clin. Invest. 75: 427-435.
97. Kase, B. F., J. I. Pedersen, K. 0. Wathne, J. Gustafsson,
and I. BjBrkhem. 1991. Importance of peroxisomes in
the formation of chenodeoxycholic acid in human
liver. Metabolism of 3a,7adihydroxy-5~cholestanoic
acid in Zellweger syndrome. Pediatr. Res. 29: 64-69.
98. Svoboda, J. A., M. J. Thompson, and W. E. Robbins.
1967. Desmosterol, an intermediate in dealkylation of
p-sitosterol in tobacco hornworm. 12ve Sci. 6: 395-404.
99. Douglass, T. S., W. E. Connor, and D. S. Lin. 1981. The
biosynthesis, absorption, apd origin of cholesterol and
plant sterols in the Florida land crab. J. Lipid Res. 22:
961-970.
100. Subbiah, M. T. R., and A. Kuksis. 1973. Differences in
metabolism of cholesterol and sitosterol following in-
travenous injection in rats. Biochim. Biophys. Acta. 306:
95-105.
101. Kritchevsky, D., L. M. Davidsson, E. H. Mosbach, and B.
I. Cohen. 1981. Identification of acid steroids in feces
of monkeys fed p-sitosterol. Lipids. 16: 77-78.
102. Salen, G., E. H. Ahrens, Jr., and S. M. Grundy. 1970.
Metabolism of psitosterol in man. J. Clin. Invest. 4 9
952-967.
103. Muri-Boberg, K., K. Einarsson and I. Bjdrkhem. 1990.
Apparent lack of conversion of sitosterol into C24-bile
acids in humans.J. Lipid Res. 31: 1083-1088.
104. Swell, L., and C. R. Treadwell. 1961. Metabolic fate of
injected ''C-phytosterols. h o c . Soc. Exp. Biol. Med. 108:
810-81 3.
105. Skrede, B., I. Bjdrkhem, 0. Bergesen, H. J. Kayden, and
S. Skrede. 1985. Studies on 5a-sitostanol: its presence in
the serum of a patient with phytosterolemia and its
biosynthesis from plant sterols in bile fistula rats.
Biochim. Biophys. Acta. 836: 368-375.
106. Muri-Boberg, K., E. Lund, J. Olund, and I. Bjdrkhem.
1990. Formation of C-21 bile acids from plant sterols in
the rat. J. Biol. Chem. 265: 7967-7975.
107. Lund, E., K. Muri-Boberg, S. Bystrom, K. Carlstrdm, J.
Olund, and I. Bjdrkhem. 1991. Formation of novel C21-
bile acids from cholesterol in the rat. J. Biol. Chem. 266:
4929-4937.
471